Probabilistic Models for the Prediction of Target Growth Interfaces of Listeria monocytogenes on Ham and Turkey Breast Products

Abstract

This study developed growth/no growth models for predicting growth boundaries of Listeria monocytogenes on ready-to-eat cured ham and uncured turkey breast slices as a function of lactic acid concentration (0% to 4%), dipping time (0 to 4 min), and storage temperature (4 to 10 °C). A 10-strain composite of L. monocytogenes was inoculated (2 to 3 log CFU/cm²) on slices, followed by dipping into lactic acid and storage in vacuum packages for up to 30 d. Total bacterial (tryptic soy agar plus 0.6% yeast extract) and L. monocytogenes (PALCAM agar) populations were determined on day 0 and at the endpoint of storage. The combinations of parameters that allowed increases in cell counts of L. monocytogenes of at least l log CFU/cm² were assigned the value of 1, while those limiting growth to <1 log CFU/cm² were given the value of 0. The binary data were used in logistic regression analysis for development of models to predict boundaries between growth and no growth of the pathogen at desired probabilities. Indices of model performance and validation with limited available data indicated that the models developed had acceptable goodness of fit. Thus, the described procedures using bacterial growth data from studies with food products may be appropriate in developing growth/no growth models to predict growth and to select lactic acid concentrations and dipping times for control of L. monocytogenes. The models developed in this study may be useful in selecting lactic acid concentrations and dipping times to control growth of Listeria monocytogenes on cured ham and uncured turkey breast during product storage, and in determining probabilities of growth under selected conditions. The modeling procedures followed may also be used for application in model development for other products, conditions, or pathogens.